CDER: An Overview and Recent Development on Business Microdata · Overview of the Linkable File...

Delivering insight through data for a better Canada

CDER: An Overview and Recent Development on Business Microdata

Jiang Beryl Li

Economic Analysis Division, Statistics Canada


Roadmap

What is CDER?

What microdata are available?

What is new?

How to access CDER?1


1. What is CDER? 2. Business microdata 3. What is new? 4. How to access?

Business Microdata: challenges

2

Identity of units must be protected• No individual or business can be directly or indirectly identified

Unlike social microdata, units in business microdata are readily identifiable even without direct identifiers

• The distribution of business data is often highly skewed (i.e., it is difficult to protect the identity of large firms)

• The sparsity of the data in certain dimensions (e.g., detailed industry and geography)

There are greater financial incentives to identifying the particulars of some firms and their competitors



Canadian Centre for Data Development and Economic Research (CDER)

3

CDER was created in 2011 to make better use of Statistics Canada’s business microdata holdings without compromising security

• Serves as a repository for business microdata

• Leads the development of new business microdata

• Provides external researchers with secure access for research-oriented projects• Five safes of access: projects; researchers; workplace; data and outputs

Institutions served• About 30 Canadian and foreign academic institutions

• 16 federal government departments and agencies, provincial government and non-profit institutions

Until recently, access was only provided at Statistics Canada’s head office in Ottawa• Application and approval process required for access

• Researchers must cover the full cost of their project



Business Microdata at CDER

4

Stand-alone, research-ready data already in use• E.g., Trade by Characteristics of Enterprises; Survey of Financing and Growth of SMEs linked to tax

data; and Annual Survey of Manufactures

Linkable File Environment (LFE)• Specific variables from a set of files where linkages have been done, but files are so large that

extractions are made upon request

Developmental datasets and other linkage environments• Analytical databases containing derived variables for specific analyses (e.g., National Accounts

Longitudinal Microdata File); additions to LFE; other linkage environments (e.g., Canadian Employer-Employee Dynamic Database); new stand-alone data



Linkable File Environment (LFE)

5

The LFE is an environment that contains datasets from administrative and surveyed sources that are linkable (the links have been done, proven and documented), but because of the size of the databases involved are not stored as one database

http://www23.statcan.gc.ca/imdb/p2SV_f.pl?Function=getSurvey&SDDS=6000

Statistics Canada’s Business Register is the “central source” of the LFE environment

Variables are extracted and “custom research dataset” are produced for specific analytical project

http://www23.statcan.gc.ca/imdb/p2SV_f.pl?Function=getSurvey&SDDS=6000



Administrative Datasets in the LFE

6

Business Register (BR), 2000-2017

Longitudinal Employment Analysis Program (LEAP), 2000-2017

General Index of Financial Information (GIFI – T1), 2005-2016

General Index of Financial Information (GIFI – T2), 2000-2017

Statement of Remuneration Paid (GIFI – T4), 2000-2016

Payroll Deduction Accounts (PD7), 2001-2017

Trade by Exporter and Importer Characteristics (TEC and TIC), 2010-2018

Patents (Canadian Intellectual Property Office), 2001-2006

United States Patent Office (USPTO - Canadian Enterprises only), 2000-2011

Horizontal Innovation and Clean Tech (HICT), 2007-2016



Survey Datasets in the LFE

7

Survey of Electronic Commerce and Technology (SECT), 2000-2007

Survey of Innovation (INNO), 2003 and 2005

Survey of Innovation and Business Strategy (SIBS), 2009, 2012, and 2016 (Summer 2019)

Survey of Advanced Technology (SAT), 2007 and 2014

Survey of Intellectual Property Management (SIPM), 2010

Survey of Financing and Growth of SMEs (SFGSME), 2007, 2011, 2014

Survey of Digital Technology and Internet Use (SDTIU), 2012 and 2013

Annual Survey of Research and Development in Canadian Industry (RDCI), 2000-2013

Foreign Direct Investment and Canadian Direct Investment Abroad (FDI and CDIA):2000-2017

Trade in Commercial Services (TICS), 2000-2014

Census of Agriculture, 2016



Overview of the Linkable File Environment

8

Census of Agriculture

SIBS

PD7 GIFI

(T1-T2-T4)

RDCI

TIC

TEC

LEAP

INNO

SATSECT

CIPO and

USPTO

TICS

FDIC and

CDIA

SFSME

Business Register

(BR)

SIPMI

SDTIUI

HICT



Trade by Enterprise Characteristics

9

Enterprise level, 2010-2018

By Industry, province, CMA, concentration and by employment of enterprise

Trade by Exporter Characteristics (TEC)• Value of exports by HS8 commodity

• By country of destination, including U.S. States breakdown

Trade by Importer Characteristics (TIC)• Value of imports by HS10 commodity

• By country of origin, country of export, and by related-parties

Uses• Trade dynamics: barriers to trade, firms’ life cycle in the export/import market

• Firm dynamics: trade orientation, owner demographics and firm characteristics (if linked to administrative data)

• Industry or product concentration: Herfindahl index



Survey of Financing and Growth of SMEs (SFGSME)

10

Cross-sectional survey• 2000, 2001, 2004, 2007, 2011, 2014 and 2017

Survey information• Use of financing during start up• Requests for financing (term loans, mortgages, lines of credit, credit cards, government loans, equity)

and outcomes (approved/rejected, collateral, term, interest rate, amounts requested/received)• Business information (exports, R&D, innovation, IP use, plans for growth, public procurement

participation)• Owner information (age, education, experience, country of birth, language of primary decision maker;

%female, %aboriginal, %visible minority)

Linked to admin. data on firm performance, 2000-2017

Uses• Financing cost; lending relations; and financial constraints on firm dynamics



Annual Survey of Manufactures (ASM)

11

ASM (1961-2012) – a series of longitudinal datasets• Cross-sectional, establishment level survey of manufacturing

• Principal industrial statistics and commodity file…100s of variables in the latest database

• Essentially a census up to 2012

ASM has been linked to other sources• National Pollutant Release Inventory (NPRI) and Greenhouse Gas Reporting Protocol (GHGRP),

plant level, 2000 to 2012

• General Index of Financial Information (GIFI), ASM-enterprise level, 2000 to 2012

• Research and Development in Canadian Industries, ASM-enterprise level, 2000 to 2009

Uses• Productivity, trade, and innovation



Longitudinal Employment Analysis Program (LEAP)

12

Firm level, 1983 to 2017

Program information• Employment, payroll, first/last year, industry and province

• Labour tracking

T2-LEAP• LEAP linked to core administrative data (including capital investment program) from the corporate

tax system covering 1983 to 2017; 1997, 2004, 2007, 2008 to 2017 vintages

Uses• Firm entry and exit

• Job creation and destruction



National Accounts Longitudinal Microdata File (NALMF)

13

Longitudinal database of Canadian enterprises, 2000-2016• Successor to T2-Longitudinal Employment Analysis Program• Tracks a richer set of firm characteristics over time (e.g., employment, payroll, revenue, profit, assets,

tangible assets, R&D, investment, and productivity)• Updated longitudinal structure

Main data sources• Statistics Canada’s Business Register; Corporation Income Tax (T2); Payroll Account Deductions (PD7)

and Statements of Remuneration Paid (T4 slip); and Goods and Services Tax (GST/HST)

Uses• GDP and employment across firm-size categories• Activities at sub-provincial levels (e.g., CMAs and ERs)• Firm dynamics, survival, and growth



Canadian Employer-Employee Dynamics Database (CEEDD)

14

Matched employer-employee database, 2001 to 2016

Covers the universe of • Individual tax filers and their families

• Employees receiving T4 slips

• Unincorporated businesses and corporations

• Unincorporated business owners reporting self-employment income

• Owners of Canadian-controlled Private Corporations

Links across various administrative tax files• T1: Personal Master File (T1PMF); Historical Personal Master File (T1H); Family File (T1FF)

• Longitudinal Immigration Database (IMDB); Temporary Residents File (TR)

• T1: Business Declaration File (T1BD); Financial Declaration File (T1FD)

• National Accounts Longitudinal Microdata File (NALMF)

• T2 Schedule 50; T4 Statement of Remuneration Paid Files (T4); Record of Employment (ROE)

• Trade by Exporter Characteristics (TEC); Trade by Importer Characteristics (TIC)



CEEDD – Linkage environment

15



Possible Uses of CEEDD

16

Labour market and firm dynamics• E.g., productivity and wage dispersion; the impact of merger and acquisition; the impact of start-

ups; entrepreneurs’ hiring decision; and entrepreneurship on a fine spatial scale, etc.

Immigrants• E.g., careers and network; initial firm allocation and earning growth; business ownership; financial

constraints and survivals of immigrant-owned firms; and trade orientation, etc.

Government programs: evaluation and policy development• E.g., employment at the economic regional level; business ownership by gender, age. immigration

status, and by marital status



CEEDD

17

Business Owner Module (2005-2016)• Immigrant

• Gender

• Age

Geographic information (2011 census boundary)• Census metropolitan areas (CMAs)

• Self-contained labour areas (SLAs)

Linkage to the CEEDD (assessment to be completed in March 2020)• Census: Indigenous group, occupation, and education, etc.

• PATSTAT: Applications, grants, patent families, citations, patent classifications, applicants, inventors, and legal status



Employment and Social Development Canada (ESDC): Request for Proposals (Apply now!)

18

A set of linked files to CEEDD to assess the labour market impacts of the Temporary Foreign Worker (TFW) Program

• ESDC: Record of Employment File; TFW Program File (2010+)• IRCC: Temporary Resident File; Landed Immigrant File

ESDC is supporting effective use of these data by issuing up to two contracts to develop a research network that will address:

• The impacts of the TFW Program on the Canadian labour market, including potential suppression of wages and displacement of Canadian workers;

• The extent to which the TFW Program is meeting the needs of Canadian firms and the labour market as a whole; and,

• The effects of the reformed TFW Program on Canadian businesses and the labour market.

Apply at https://buyandsell.gc.ca/tma/node/875503• Closing date is September 30, 2019

https://buyandsell.gc.ca/tma/node/875503



New Business Microdata at CDER

19

International Accounts and Trade• Canada and the World Statistics Hub (13-609X): Updated on an ongoing basis

• Trade by Enterprises Characteristics: 2010-2018 (May 2019)

• Trade by Enterprises Characteristics, by census metropolitan area: 2010-2018 (Sep 2019)

• Foreign Direct Investment and Canadian Direct Investment Aboard: 2000-2018 (Apr 2019)

• Outward Foreign Affiliates Statistics (FATS): 2011-2016 (RY2017 in fall 2019)

• Enhanced details on securities statistics: 2015Q1 to 2019Q1

• Inward FATS and activities of multinational firms in Canada (RY2010-2017 in Nov 2019)

• Canada’s international trade in services: 2017 (Oct 2018)



New Business Microdata at CDER (cont’d)

20

Investment, Science, and Technology• Canadian Survey of Cyber Security and Cybercrime (RY2017)

• Survey of Innovation and Business Strategy (RY2017 in Jun 2019)

• Innovation

• Advanced technology use

• Business strategies and practices

• Global value chains

• Canadian Housing Statistics Program (RY2017-2018)

• Link numerous administrative files (e.g., tax files, property assessment files, etc.)

• Used to understand and address foreign-home-ownership-related issues

• By geography(CMAs & CSDs in Vancouver and Toronto), property/owner characteristics, property use/financing



New Business Microdata at CDER (cont’d)

21

Farm-level data• Census of Agriculture: 1986, 1991, 1996, 2001, 2006, 2011 and 2016 (longitudinal files underlined)

• Farm Financial Survey, 2013, 2015 and 2017

• Livestock, crops and horticulture (potential)

Price Indices• Cover a range of goods, services and construction outputs

• By type: transaction price; unit value; margin price; list price; and model price

• Unpublished indices are available in many cases (incl. by geography/product/industry) and with a more recent starting date


1. What is CDER? 2. Business microdata 3. What is New? 4. How to access?

Application Process

22

By type of researcher: government (federal/provincial) and non-government

For non-government researchers, the process is as follows1) Contact CDER and draft a proposal2) Submit final proposal 3) Proposal evaluation: peer review and approved by an internal Statistics Canada committee4) Undergo a security screening5) Take the Oath of the Office of the Statistics Act to become deemed employee6) Sign a Microdata Research Contract

Contact:• Website: http://www.statcan.gc.ca/cder• Email: [email protected]

http://www.statcan.gc.ca/cder

mailto:[email protected]



Improving Access to Business Microdata

1 – Pilot Projects

23

T2-Longitudinal Employment Program (2001-2015): April 2018

Survey of Financing and Growth of Small and Medium-sized Enterprises 2014 linked to administrative data (2010-2014): November 2018

Project setup• Calls for proposal and webinars

• Approved research projects that focus on multivariate analysis are being carried out in the RDCs in 2018-2019 and 2019-2020

• Researchers develop their models using the synthetic/treated data at the RDCs

• Programs are then sent to CDER by RDC analysts to be run on the actual data

• Final results are released from CDER by e-mail

Ultimate objectives• From the experience gained, similar datasets can be placed in the RDCs



Improving Access to Business Microdata (Cont’d)

2 – RDC Releases

24

Core business microdata (March 2020)• An LFE extraction: A selection of administrative and survey data

• The variable list is to be finalized after consultations with internal/external stakeholders

Core Canadian Employer-Employee Dynamic Database (March 2021)• Stay tuned!



Improving Access to Business Microdata (Cont’d)

3 – Productivity Partnership

25

Team of experts from the academic, private, and public sectors

Lowers barriers to access business data for academics and students by providing funding for productivity research projects

Statistics Canada liaison researcher: Beryl Li, [email protected]

Important partner in the development of synthetic datasets

Collaboration and support to clarify the CDER application process steps

https://productivitypartnership.ca

https://productivitypartnership.ca/

Thank You!

For more information, please visit

www.statcan.gc.ca/eng/cder/index

26

CDER: An Overview and Recent Development on Business Microdata · Overview of the Linkable File...

Documents

Transcript of CDER: An Overview and Recent Development on Business Microdata · Overview of the Linkable File...